use of uk.ac.babraham.SeqMonk.DataTypes.Genome.Location in project SeqMonk by s-andrews.
the class FeaturePositionSelectorPanel method getCoreProbes.
/**
* Gets the set of locations for the core of each feature. This wouldn't
* include additional context added by the options, but would have subtracted
* context removed by the options.
*
* @return
*/
public Probe[] getCoreProbes() {
Chromosome[] chromosomes = collection.genome().getAllChromosomes();
Vector<Probe> newProbes = new Vector<Probe>();
for (int c = 0; c < chromosomes.length; c++) {
Vector<Feature> allFeatures = new Vector<Feature>();
String[] selectedFeatureTypes = selectedFeatureTypes();
for (int f = 0; f < selectedFeatureTypes.length; f++) {
Feature[] features = collection.genome().annotationCollection().getFeaturesForType(chromosomes[c], selectedFeatureTypes[f]);
for (int i = 0; i < features.length; i++) {
allFeatures.add(features[i]);
}
}
Feature[] features = allFeatures.toArray(new Feature[0]);
for (int f = 0; f < features.length; f++) {
if (useSubFeatures()) {
// We need to split this up so get the sub-features
if (features[f].location() instanceof SplitLocation) {
SplitLocation location = (SplitLocation) features[f].location();
Location[] subLocations = location.subLocations();
if (useExonSubfeatures()) {
for (int s = 0; s < subLocations.length; s++) {
makeProbes(features[f], chromosomes[c], subLocations[s], newProbes, true);
}
} else {
// We're making introns
for (int s = 1; s < subLocations.length; s++) {
makeProbes(features[f], chromosomes[c], new Location(subLocations[s - 1].end() + 1, subLocations[s].start() - 1, features[f].location().strand()), newProbes, true);
}
}
} else {
if (useExonSubfeatures()) {
// We can still make a single probe
makeProbes(features[f], chromosomes[c], features[f].location(), newProbes, true);
}
// If we're making introns then we're stuffed and we give up.
}
} else {
makeProbes(features[f], chromosomes[c], features[f].location(), newProbes, true);
}
}
}
Probe[] finalList = newProbes.toArray(new Probe[0]);
if (removeDuplicates()) {
finalList = removeDuplicates(finalList);
}
return finalList;
}
use of uk.ac.babraham.SeqMonk.DataTypes.Genome.Location in project SeqMonk by s-andrews.
the class GFF3AnnotationParser method parseAnnotation.
public AnnotationSet[] parseAnnotation(File file, Genome genome, String prefix) throws Exception {
System.err.println("Parsing " + file);
if (prefix == null) {
featurePrefix = JOptionPane.showInputDialog(SeqMonkApplication.getInstance(), "Feature prefix", "GFFv3/GTP Options", JOptionPane.QUESTION_MESSAGE);
} else {
featurePrefix = prefix;
}
if (featurePrefix == null)
featurePrefix = "";
Vector<AnnotationSet> annotationSets = new Vector<AnnotationSet>();
AnnotationSet currentAnnotation = new AnnotationSet(genome, file.getName());
annotationSets.add(currentAnnotation);
Hashtable<String, FeatureGroup> groupedFeatures = new Hashtable<String, FeatureGroup>();
BufferedReader br;
if (file.getName().toLowerCase().endsWith(".gz")) {
br = new BufferedReader(new InputStreamReader(new GZIPInputStream(new FileInputStream(file))));
} else {
br = new BufferedReader(new FileReader(file));
}
String line;
int count = 0;
while ((line = br.readLine()) != null) {
if (cancel) {
progressCancelled();
br.close();
return null;
}
if (count % 1000 == 0) {
progressUpdated("Read " + count + " lines from " + file.getName(), 0, 1);
}
if (count > 1000000 && count % 1000000 == 0) {
progressUpdated("Caching...", 0, 1);
currentAnnotation.finalise();
currentAnnotation = new AnnotationSet(genome, file.getName() + "[" + annotationSets.size() + "]");
annotationSets.add(currentAnnotation);
}
++count;
// Ignore blank lines
if (line.trim().length() == 0)
continue;
// Skip comments
if (line.startsWith("#"))
continue;
String[] sections = line.split("\t");
// Check to see if we've got enough data to work with
if (sections.length < 7) {
progressWarningReceived(new SeqMonkException("Not enough data from line '" + line + "'"));
continue;
}
int strand;
int start;
int end;
try {
start = Integer.parseInt(sections[3]);
end = Integer.parseInt(sections[4]);
// End must always be later than start
if (end < start) {
int temp = start;
start = end;
end = temp;
}
if (sections.length >= 7) {
if (sections[6].equals("+")) {
strand = Location.FORWARD;
} else if (sections[6].equals("-")) {
strand = Location.REVERSE;
} else {
strand = Location.UNKNOWN;
}
} else {
strand = Location.UNKNOWN;
}
} catch (NumberFormatException e) {
progressWarningReceived(new SeqMonkException("Location " + sections[3] + "-" + sections[4] + " was not an integer"));
continue;
}
ChromosomeWithOffset c;
try {
c = genome.getChromosome(sections[0]);
} catch (IllegalArgumentException e) {
progressWarningReceived(new SeqMonkException("Couldn't find a chromosome called " + sections[0]));
continue;
}
start = c.position(start);
end = c.position(end);
// We also don't allow readings which are beyond the end of the chromosome
if (end > c.chromosome().length()) {
int overrun = end - c.chromosome().length();
progressWarningReceived(new SeqMonkException("Reading position " + end + " was " + overrun + "bp beyond the end of chr" + c.chromosome().name() + " (" + c.chromosome().length() + ")"));
continue;
}
if (sections.length > 8 && sections[8].trim().length() > 0) {
// Should check for escaped colons
String[] attributes = sections[8].split(" *; *");
// Make up a data structure of the attributes we have
Hashtable<String, Vector<String>> keyValuePairs = new Hashtable<String, Vector<String>>();
for (int a = 0; a < attributes.length; a++) {
// Should check for escaped equals
String[] keyValue = attributes[a].split("=", 2);
// See if we didn't get split
if (keyValue.length == 1) {
// This could be a GTF file which uses quoted values in space delimited fields
keyValue = attributes[a].split(" \"");
if (keyValue.length == 2) {
// We need to remove the quote from the end of the value
keyValue[1] = keyValue[1].substring(0, keyValue[1].length() - 1);
// System.out.println("Key='"+keyValue[0]+"' value='"+keyValue[1]+"'");
}
}
if (keyValue.length == 2) {
if (keyValuePairs.containsKey(keyValue[0])) {
keyValuePairs.get(keyValue[0]).add(keyValue[1]);
} else {
Vector<String> newVector = new Vector<String>();
newVector.add(keyValue[1]);
keyValuePairs.put(keyValue[0], newVector);
}
} else {
progressWarningReceived(new SeqMonkException("No key value delimiter in " + attributes[a]));
}
}
if (keyValuePairs.containsKey("Parent") && !sections[2].equals("mRNA")) {
// We change exons to mRNA so we don't end up with spliced exon objects
if (sections[2].equals("exon"))
sections[2] = "mRNA";
String[] parents = keyValuePairs.get("Parent").elementAt(0).split(",");
for (int p = 0; p < parents.length; p++) {
if (!groupedFeatures.containsKey(sections[2] + "_" + parents[p])) {
// Make a new feature to which we can add this
Feature feature = new Feature(featurePrefix + sections[2], c.chromosome().name());
groupedFeatures.put(sections[2] + "_" + parents[p], new FeatureGroup(feature, strand, feature.location()));
Enumeration<String> en = keyValuePairs.keys();
while (en.hasMoreElements()) {
String key = en.nextElement();
String[] values = keyValuePairs.get(key).toArray(new String[0]);
for (int v = 0; v < values.length; v++) {
feature.addAttribute(key, values[v]);
}
}
}
groupedFeatures.get(sections[2] + "_" + parents[p]).addSublocation(new Location(start, end, strand));
}
} else // parent feature
if (keyValuePairs.containsKey("transcript_id")) {
if (sections[2].equals("exon"))
sections[2] = "mRNA";
if (!groupedFeatures.containsKey(sections[2] + "_" + keyValuePairs.get("transcript_id").elementAt(0))) {
Feature feature = new Feature(featurePrefix + sections[2], c.chromosome().name());
Enumeration<String> en = keyValuePairs.keys();
while (en.hasMoreElements()) {
String key = en.nextElement();
String[] values = keyValuePairs.get(key).toArray(new String[0]);
for (int v = 0; v < values.length; v++) {
feature.addAttribute(key, values[v]);
}
}
groupedFeatures.put(sections[2] + "_" + keyValuePairs.get("transcript_id").elementAt(0), new FeatureGroup(feature, strand, feature.location()));
}
groupedFeatures.get(sections[2] + "_" + keyValuePairs.get("transcript_id").elementAt(0)).addSublocation(new Location(start, end, strand));
} else {
// If we get here we're making a feature with attributes
Feature feature = new Feature(featurePrefix + sections[2], c.chromosome().name());
feature.setLocation(new Location(start, end, strand));
Enumeration<String> en = keyValuePairs.keys();
while (en.hasMoreElements()) {
String key = en.nextElement();
String[] values = keyValuePairs.get(key).toArray(new String[0]);
for (int v = 0; v < values.length; v++) {
feature.addAttribute(key, values[v]);
}
}
if (keyValuePairs.containsKey("ID")) {
// This is a feature which may end up having subfeatures
groupedFeatures.put(sections[2] + "_" + keyValuePairs.get("ID").elementAt(0), new FeatureGroup(feature, strand, feature.location()));
// System.out.println("Making new entry for "+keyValuePairs.get("ID").elementAt(0));
} else {
// We can just add this to the annotation collection
currentAnnotation.addFeature(feature);
}
}
} else {
// No group parameter to worry about
Feature feature = new Feature(featurePrefix + sections[2], c.chromosome().name());
feature.setLocation(new Location(start, end, strand));
currentAnnotation.addFeature(feature);
}
}
br.close();
// Now go through the grouped features adding them to the annotation set
Iterator<FeatureGroup> i = groupedFeatures.values().iterator();
while (i.hasNext()) {
Feature f = i.next().feature();
currentAnnotation.addFeature(f);
}
return annotationSets.toArray(new AnnotationSet[0]);
}
use of uk.ac.babraham.SeqMonk.DataTypes.Genome.Location in project SeqMonk by s-andrews.
the class ProbeListAnnotationParser method parseAnnotation.
/* (non-Javadoc)
* @see uk.ac.babraham.SeqMonk.AnnotationParsers.AnnotationParser#parseAnnotation(java.io.File, uk.ac.babraham.SeqMonk.DataTypes.Genome.Genome)
*/
protected AnnotationSet[] parseAnnotation(File file, Genome genome) throws Exception {
Vector<AnnotationSet> annotationSets = new Vector<AnnotationSet>();
AnnotationSet currentAnnotation = new AnnotationSet(genome, probeList.name());
annotationSets.add(currentAnnotation);
Probe[] probes = probeList.getAllProbes();
for (int p = 0; p < probes.length; p++) {
if (p % 1 + (probes.length / 100) == 0) {
progressUpdated("Converted " + p + " probes", p, probes.length);
}
if (p > 1000000 && p % 1000000 == 0) {
progressUpdated("Caching...", 0, 1);
currentAnnotation.finalise();
currentAnnotation = new AnnotationSet(genome, probeList.name() + "[" + annotationSets.size() + "]");
annotationSets.add(currentAnnotation);
}
Feature feature = new Feature(featureType, probes[p].chromosome().name());
if (probes[p].hasDefinedName()) {
feature.addAttribute("name", probes[p].name());
}
feature.setLocation(new Location(probes[p].start(), probes[p].end(), probes[p].strand()));
currentAnnotation.addFeature(feature);
}
return annotationSets.toArray(new AnnotationSet[0]);
}
use of uk.ac.babraham.SeqMonk.DataTypes.Genome.Location in project SeqMonk by s-andrews.
the class FeaturePercentileProbeGenerator method run.
/* (non-Javadoc)
* @see java.lang.Runnable#run()
*/
public void run() {
Chromosome[] chromosomes = collection.genome().getAllChromosomes();
Vector<Probe> newProbes = new Vector<Probe>();
for (int c = 0; c < chromosomes.length; c++) {
// Time for an update
updateGenerationProgress("Processed " + c + " chromosomes", c, chromosomes.length);
Feature[] features = collection.genome().annotationCollection().getFeaturesForType(chromosomes[c], featureType);
for (int f = 0; f < features.length; f++) {
// See if we need to quit
if (cancel) {
generationCancelled();
return;
}
if (useSubfeatures && (features[f].location() instanceof SplitLocation)) {
SplitLocation location = (SplitLocation) features[f].location();
Location[] subLocations = location.subLocations();
for (int s = 0; s < subLocations.length; s++) {
makeProbes(features[f], chromosomes[c], subLocations[s], newProbes);
}
} else {
makeProbes(features[f], chromosomes[c], features[f].location(), newProbes);
}
}
}
Probe[] finalList = newProbes.toArray(new Probe[0]);
ProbeSet finalSet = new ProbeSet(getDescription(), finalList);
generationComplete(finalSet);
}
use of uk.ac.babraham.SeqMonk.DataTypes.Genome.Location in project SeqMonk by s-andrews.
the class FeatureGroup method getSubLocations.
public Location[] getSubLocations() {
if (features.size() == 1) {
Location loc = features.elementAt(0).location();
if (loc instanceof SplitLocation) {
return ((SplitLocation) loc).subLocations();
} else {
return new Location[] { loc };
}
}
LongVector allLocs = new LongVector();
Enumeration<Feature> en = features.elements();
while (en.hasMoreElements()) {
Location loc = en.nextElement().location();
if (loc instanceof SplitLocation) {
Location[] subLocs = ((SplitLocation) loc).subLocations();
for (int s = 0; s < subLocs.length; s++) {
allLocs.add(subLocs[s].packedPosition());
}
} else {
allLocs.add(loc.packedPosition());
}
}
long[] locs = allLocs.toArray();
SequenceRead.sort(locs);
Vector<Location> mergedLocs = new Vector<Location>();
long current = locs[0];
for (int i = 1; i < locs.length; i++) {
// if (debug) {System.err.println("Looking at "+SequenceRead.start(locs[i])+"-"+SequenceRead.end(locs[i])+" current is "+SequenceRead.start(current)+"-"+SequenceRead.end(current));}
if (SequenceRead.overlaps(current, locs[i]) && SequenceRead.end(locs[i]) > SequenceRead.end(current)) {
// if (debug) {System.err.println("They overlap, extending...");}
current = SequenceRead.packPosition(SequenceRead.start(current), SequenceRead.end(locs[i]), SequenceRead.strand(current));
} else if (SequenceRead.end(locs[i]) <= SequenceRead.end(current)) {
// if (debug) {System.err.println("This is a subset, ignoring it");}
continue;
} else {
// if (debug) {System.err.println("They don't overlap, moving on...");}
mergedLocs.add(new Location(current));
current = locs[i];
}
}
mergedLocs.add(new Location(current));
return mergedLocs.toArray(new Location[0]);
}
Aggregations